SciBet as a portable and fast single cell type identifier

Single substitution in H3.3G34 alters DNMT3A recruitment to cause progressive neurodegeneration

Germline histone H3.3 amino acid substitutions, including H3.3G34R/V, cause severe neurodevelopmental syndromes. To understand how these mutations impact brain development, we generated H3.3G34R/V/W knock-in mice and identified strikingly distinct developmental defects for each mutation. H3.3G34R-mutants exhibited progressive microcephaly and neurodegeneration, with abnormal accumulation of disease-associated microglia and concurrent neuronal depletion. G34R severely decreased H3K36me2 on the mutant H3.3 tail, impairing recruitment of DNA methyltransferase DNMT3A and its redistribution on chromatin. These changes were concurrent with sustained expression of complement and other innate immune genes possibly through loss of non-CG (CH) methylation and silencing of neuronal gene promoters through aberrant CG methylation. Complement expression in G34R brains may lead to neuroinflammation possibly accounting for progressive neurodegeneration. Our study reveals that H3.3G34-substitutions have differential impact on the epigenome, which underlie the diverse phenotypes observed, and uncovers potential roles for H3K36me2 and DNMT3A-dependent CH-methylation in modulating synaptic pruning and neuroinflammation in post-natal brains.

Discovery of Targets for Immune-Metabolic Antitumor Drugs Identifies Estrogen-Related Receptor Alpha

Drugs that kill tumors through multiple mechanisms have the potential for broad clinical benefits. Here, we first developed an in silico multiomics approach (BipotentR) to find cancer cell-specific regulators that simultaneously modulate tumor immunity and another oncogenic pathway and then used it to identify 38 candidate immune-metabolic regulators. We show the tumor activities of these regulators stratify patients with melanoma by their response to anti-PD-1 using machine learning and deep neural approaches, which improve the predictive power of current biomarkers. The topmost identified regulator, ESRRA, is activated in immunotherapy-resistant tumors. Its inhibition killed tumors by suppressing energy metabolism and activating two immune mechanisms: (i) cytokine induction, causing proinflammatory macrophage polarization, and (ii) antigen-presentation stimulation, recruiting CD8+ T cells into tumors. We also demonstrate a wide utility of BipotentR by applying it to angiogenesis and growth suppressor evasion pathways. BipotentR (http://bipotentr.dfci.harvard.edu/) provides a resource for evaluating patient response and discovering drug targets that act simultaneously through multiple mechanisms.

SIGNIFICANCE

BipotentR presents resources for evaluating patient response and identifying targets for drugs that can kill tumors through multiple mechanisms concurrently. Inhibition of the topmost candidate target killed tumors by suppressing energy metabolism and effects on two immune mechanisms. This article is highlighted in the In This Issue feature, p. 517.

A population of stem cells with strong regenerative potential discovered in deer antlers

The annual regrowth of deer antlers provides a valuable model for studying organ regeneration in mammals. We describe a single-cell atlas of antler regrowth. The earliest-stage antler initiators were mesenchymal cells that express the paired related homeobox 1 gene (PRRX1+ mesenchymal cells). We also identified a population of "antler blastema progenitor cells" (ABPCs) that developed from the PRRX1+ mesenchymal cells and directed the antler regeneration process. Cross-species comparisons identified ABPCs in several mammalian blastema. In vivo and in vitro ABPCs displayed strong self-renewal ability and could generate osteochondral lineage cells. Last, we observed a spatially well-structured pattern of cellular and gene expression in antler growth center during the peak growth stage, revealing the cellular mechanisms involved in rapid antler elongation.

Immune checkpoint HLA-E:CD94-NKG2A mediates evasion of circulating tumor cells from NK cell surveillance

Circulating tumor cells (CTCs), shed by primary malignancies, function as "seeds" for distant metastasis. However, it is still largely unknown how CTCs escape immune surveillance. Here, we characterize the transcriptomes of human pancreatic ductal adenocarcinoma CTCs, primary, and metastatic lesions at single-cell scale. Cell-interaction analysis and functional studies in vitro and in vivo reveal that CTCs and natural killer (NK) cells interact via the immune checkpoint molecule pair HLA-E:CD94-NKG2A. Disruption of this interaction by blockade of NKG2A or knockdown of HLA-E expression enhances NK-mediated tumor cell killing in vitro and prevents tumor metastasis in vivo. Mechanistic studies indicate that platelet-derived RGS18 promotes the expression of HLA-E through AKT-GSK3β-CREB signaling, and overexpression of RGS18 facilitates pancreatic tumor hepatic metastasis. In conclusion, platelet-derived RGS18 protects CTCs from NK-mediated immune surveillance by engaging the immune checkpoint HLA-E:CD94-NKG2A. Interruption of the suppressive signaling prevents tumor metastasis in vivo by immune elimination of CTCs.

Transformer for one stop interpretable cell type annotation

Consistent annotation transfer from reference dataset to query dataset is fundamental to the development and reproducibility of single-cell research. Compared with traditional annotation methods, deep learning based methods are faster and more automated. A series of useful single cell analysis tools based on autoencoder architecture have been developed but these struggle to strike a balance between depth and interpretability. Here, we present TOSICA, a multi-head self-attention deep learning model based on Transformer that enables interpretable cell type annotation using biologically understandable entities, such as pathways or regulons. We show that TOSICA achieves fast and accurate one-stop annotation and batch-insensitive integration while providing biologically interpretable insights for understanding cellular behavior during development and disease progressions. We demonstrate TOSICA's advantages by applying it to scRNA-seq data of tumor-infiltrating immune cells, and CD14+ monocytes in COVID-19 to reveal rare cell types, heterogeneity and dynamic trajectories associated with disease progression and severity.

Cross-species cell-type assignment from single-cell RNA-seq data by a heterogeneous graph neural network

Cross-species comparative analyses of single-cell RNA sequencing (scRNA-seq) data allow us to explore, at single-cell resolution, the origins of the cellular diversity and evolutionary mechanisms that shape cellular form and function. Cell-type assignment is a crucial step to achieve that. However, the poorly annotated genome and limited known biomarkers hinder us from assigning cell identities for nonmodel species. Here, we design a heterogeneous graph neural network model, CAME, to learn aligned and interpretable cell and gene embeddings for cross-species cell-type assignment and gene module extraction from scRNA-seq data. CAME achieves significant improvements in cell-type characterization across distant species owing to the utilization of non-one-to-one homologous gene mapping ignored by early methods. Our large-scale benchmarking study shows that CAME significantly outperforms five classical methods in terms of cell-type assignment and model robustness to insufficiency and inconsistency of sequencing depths. CAME can transfer the major cell types and interneuron subtypes of human brains to mouse and discover shared cell-type-specific functions in homologous gene modules. CAME can align the trajectories of human and macaque spermatogenesis and reveal their conservative expression dynamics. In short, CAME can make accurate cross-species cell-type assignments even for nonmodel species and uncover shared and divergent characteristics between two species from scRNA-seq data.

scWECTA: A weighted ensemble classification framework for cell type assignment based on single cell transcriptome

Rapid advances in single-cell transcriptome analysis provide deeper insights into the study of tissue heterogeneity at the cellular level. Unsupervised clustering can identify potential cell populations in single-cell RNA-sequencing (scRNA-seq) data, but fail to further determine the identity of each cell. Existing automatic annotation methods using scRNA-seq data based on machine learning mainly use single feature set and single classifier. In view of this, we propose a Weighted Ensemble classification framework for Cell Type Annotation, named scWECTA, which improves the accuracy of cell type identification. scWECTA uses five informative gene sets and integrates five classifiers based on soft weighted ensemble framework. And the ensemble weights are inferred through the constrained non-negative least squares. Validated on multiple pairs of scRNA-seq datasets, scWECTA is able to accurately annotate scRNA-seq data across platforms and across tissues, especially for imbalanced data containing rare cell types. Moreover, scWECTA outperforms other comparable methods in balancing the prediction accuracy of common cell types and the unassigned rate of non-common cell types at the same time. The source code of scWECTA is freely available at https://github.com/ttren-sc/scWECTA.

Privacy-preserving integration of multiple institutional data for single-cell type identification with scPrivacy

The rapid accumulation of large-scale single-cell RNA-seq datasets from multiple institutions presents remarkable opportunities for automatically cell annotations through integrative analyses. However, the privacy issue has existed but being ignored, since we are limited to access and utilize all the reference datasets distributed in different institutions globally due to the prohibited data transmission across institutions by data regulation laws. To this end, we present scPrivacy, which is the first and generalized automatically single-cell type identification prototype to facilitate single cell annotations in a data privacy-preserving collaboration manner. We evaluated scPrivacy on a comprehensive set of publicly available benchmark datasets for single-cell type identification to stimulate the scenario that the reference datasets are rapidly generated and distributed in multiple institutions, while they are prohibited to be integrated directly or exposed to each other due to the data privacy regulations, demonstrating its effectiveness, time efficiency and robustness for privacy-preserving integration of multiple institutional datasets in single cell annotations.

Spatial-ID: a cell typing method for spatially resolved transcriptomics via transfer learning and spatial embedding

Spatially resolved transcriptomics provides the opportunity to investigate the gene expression profiles and the spatial context of cells in naive state, but at low transcript detection sensitivity or with limited gene throughput. Comprehensive annotating of cell types in spatially resolved transcriptomics to understand biological processes at the single cell level remains challenging. Here we propose Spatial-ID, a supervision-based cell typing method, that combines the existing knowledge of reference single-cell RNA-seq data and the spatial information of spatially resolved transcriptomics data. We present a series of benchmarking analyses on publicly available spatially resolved transcriptomics datasets, that demonstrate the superiority of Spatial-ID compared with state-of-the-art methods. Besides, we apply Spatial-ID on a self-collected mouse brain hemisphere dataset measured by Stereo-seq, that shows the scalability of Spatial-ID to three-dimensional large field tissues with subcellular spatial resolution.

K27M in canonical and noncanonical H3 variants occurs in distinct oligodendroglial cell lineages in brain midline gliomas

Canonical (H3.1/H3.2) and noncanonical (H3.3) histone 3 K27M-mutant gliomas have unique spatiotemporal distributions, partner alterations and molecular profiles. The contribution of the cell of origin to these differences has been challenging to uncouple from the oncogenic reprogramming induced by the mutation. Here, we perform an integrated analysis of 116 tumors, including single-cell transcriptome and chromatin accessibility, 3D chromatin architecture and epigenomic profiles, and show that K27M-mutant gliomas faithfully maintain chromatin configuration at developmental genes consistent with anatomically distinct oligodendrocyte precursor cells (OPCs). H3.3K27M thalamic gliomas map to prosomere 2-derived lineages. In turn, H3.1K27M ACVR1-mutant pontine gliomas uniformly mirror early ventral NKX6-1⁺/SHH-dependent brainstem OPCs, whereas H3.3K27M gliomas frequently resemble dorsal PAX3⁺/BMP-dependent progenitors. Our data suggest a context-specific vulnerability in H3.1K27M-mutant SHH-dependent ventral OPCs, which rely on acquisition of ACVR1 mutations to drive aberrant BMP signaling required for oncogenesis. The unifying action of K27M mutations is to restrict H3K27me3 at PRC2 landing sites, whereas other epigenetic changes are mainly contingent on the cell of origin chromatin state and cycling rate.

Single-cell dissection of the obesity-exercise axis in adipose-muscle tissues implies a critical role for mesenchymal stem cells

Exercise training is critical for the prevention and treatment of obesity, but its underlying mechanisms remain incompletely understood given the challenge of profiling heterogeneous effects across multiple tissues and cell types. Here, we address this challenge and opposing effects of exercise and high-fat diet (HFD)-induced obesity at single-cell resolution in subcutaneous and visceral white adipose tissue and skeletal muscle in mice with diet and exercise training interventions. We identify a prominent role of mesenchymal stem cells (MSCs) in obesity and exercise-induced tissue adaptation. Among the pathways regulated by exercise and HFD in MSCs across the three tissues, extracellular matrix remodeling and circadian rhythm are the most prominent. Inferred cell-cell interactions implicate within- and multi-tissue crosstalk centered around MSCs. Overall, our work reveals the intricacies and diversity of multi-tissue molecular responses to exercise and obesity and uncovers a previously underappreciated role of MSCs in tissue-specific and multi-tissue beneficial effects of exercise.

PSMG2-controlled proteasome-autophagy balance mediates the tolerance for MEK-targeted therapy in triple-negative breast cancer

Although the MAPK pathway is aberrantly activated in triple-negative breast cancers (TNBCs), the clinical outcome of MEK-targeted therapy is still poor. Through a genome-wide CRISPR-Cas9 library screening, we find that inhibition of PSMG2 sensitizes TNBC cells BT549 and MB468 to the MEK inhibitor AZD6244. Mechanistically, PSMG2 knockdown impairs proteasome function, which in turn activates autophagy-mediated PDPK1 degradation. The PDPK1 degradation significantly enhances AZD6244-induced tumor cell growth inhibition by interrupting the negative feedback signals toward the AKT pathway. Consistently, co-targeting proteasomes and MEK with inhibitors synergistically suppresses tumor cell growth. The autophagy inhibitor chloroquine partially relieves the PDPK1 degradation and reverses the growth inhibition induced by combinatorial inhibition of MEK and proteasome. The combination regimen with the proteasome inhibitor MG132 plus AZD6244 synergistically inhibits tumor growth in a 4T1 xenograft mouse model. In summary, our study not only unravels the mechanism of MEK inhibitor resistance but also provides a combinatorial therapeutic strategy for TNBC in clinics.

Single-cell meta-analyses reveal responses of tumor-reactive CXCL13+ T cells to immune-checkpoint blockade

Immune-checkpoint blockade (ICB) therapies represent a paradigm shift in the treatment of human cancers; however, it remains incompletely understood how tumor-reactive T cells respond to ICB across tumor types. Here, we demonstrate that measuring CXCL13 expression could effectively identify both precursor and terminally differentiated tumor-reactive CD8⁺ T cells within tumors. Applying this approach, we performed meta-analyses of published single-cell data for CXCL13⁺CD8⁺ T cells in 225 samples from 102 patients treated with ICB across five cancer types. We found that CXCL13⁺CD8⁺ T cells were correlated with favorable responses to ICB, and the treatment further increased such cells in responsive tumors. In addition, CXCL13⁺ tumor-reactive subsets exhibited variable responses to ICB in distinct contexts, likely due to different degrees of exhaustion-related immunosuppression. Our integrated analyses provide insights into mechanisms underlying ICB and suggest that bolstering precursor tumor-reactive CD8⁺ T cells might provide an effective therapeutic approach to improve cancer treatment.

SAREV: A review on statistical analytics of single-cell RNA sequencing data

Due to the development of next-generation RNA sequencing (NGS) technologies, there has been tremendous progress in research involving determining the role of genomics, transcriptomics and epigenomics in complex biological systems. However, scientists have realized that information obtained using earlier technology, frequently called 'bulk RNA-seq' data, provides information averaged across all the cells present in a tissue. Relatively newly developed single cell (scRNA-seq) technology allows us to provide transcriptomic information at a single-cell resolution. Nevertheless, these high-resolution data have their own complex natures and demand novel statistical data analysis methods to provide effective and highly accurate results on complex biological systems. In this review, we cover many such recently developed statistical methods for researchers wanting to pursue scRNA-seq statistical and computational research as well as scientific research about these existing methods and free software tools available for their generated data. This review is certainly not exhaustive due to page limitations. We have tried to cover the popular methods starting from quality control to the downstream analysis of finding differentially expressed genes and concluding with a brief description of network analysis.

Chord: an ensemble machine learning algorithm to identify doublets in single-cell RNA sequencing data

High-throughput single-cell RNA sequencing (scRNA-seq) is a popular method, but it is accompanied by doublet rate problems that disturb the downstream analysis. Several computational approaches have been developed to detect doublets. However, most of these methods may yield satisfactory performance in some datasets but lack stability in others; thus, it is difficult to regard a single method as the gold standard which can be applied to all types of scenarios. It is a difficult and time-consuming task for researchers to choose the most appropriate software. We here propose Chord which implements a machine learning algorithm that integrates multiple doublet detection methods to address these issues. Chord had higher accuracy and stability than the individual approaches on different datasets containing real and synthetic data. Moreover, Chord was designed with a modular architecture port, which has high flexibility and adaptability to the incorporation of any new tools. Chord is a general solution to the doublet detection problem.

For the unmet need to choose the suitable doublet detection method, an ensemble machine learning algorithm called Chord was developed, which integrates multiple methods and achieves higher accuracy and stability on different scRNA-seq datasets.

scMAGIC: accurately annotating single cells using two rounds of reference-based classification

Here, we introduce scMAGIC (Single Cell annotation using MArker Genes Identification and two rounds of reference-based Classification [RBC]), a novel method that uses well-annotated single-cell RNA sequencing (scRNA-seq) data as the reference to assist in the classification of query scRNA-seq data. A key innovation in scMAGIC is the introduction of a second-round RBC in which those query cells whose cell identities are confidently validated in the first round are used as a new reference to again classify query cells, therefore eliminating the batch effects between the reference and the query data. scMAGIC significantly outperforms 13 competing RBC methods with their optimal parameter settings across 86 benchmark tests, especially when the cell types in the query dataset are not completely covered by the reference dataset and when there exist significant batch effects between the reference and the query datasets. Moreover, when no reference dataset is available, scMAGIC can annotate query cells with reasonably high accuracy by using an atlas dataset as the reference.

Single-cell analysis reveals Comma-1D as a unique cell model for mammary gland development and breast cancer

The mammary epithelial tree contains two distinct populations, luminal and basal. The investigation of how this heterogeneity is developed and how it influences tumorigenesis has been hampered by the need to perform these studies using animal models. Comma-1D is an immortalized mouse mammary epithelial cell line that has unique morphogenetic properties. By performing single-cell RNA-seq studies we found that Comma-1D cultures consist of two main populations with luminal and basal features and a smaller population with mixed lineage and bipotent characteristics. We demonstrated that multiple transcription factors associated with the differentiation of the mammary epithelium in vivo also modulate this process in Comma-1D cultures. Additionally, we found that only cells with luminal features were able to acquire transformed characteristics after an oncogenic HER2 mutant was introduced in their genomes. Overall, our studies characterize at a single-cell level the heterogeneity of the Comma-1D cell line and illustrate how Comma-1D cells can be used as an experimental model to study both the differentiation and the transformation processes in vitro.

Stearoyl-CoA Desaturase inhibition reverses immune, synaptic and cognitive impairments in an Alzheimer's disease mouse model

The defining features of Alzheimer’s disease (AD) include alterations in protein aggregation, immunity, lipid metabolism, synapses, and learning and memory. Of these, lipid abnormalities are the least understood. Here, we investigate the role of Stearoyl-CoA desaturase (SCD), a crucial regulator of fatty acid desaturation, in AD pathogenesis. We show that inhibiting brain SCD activity for 1-month in the 3xTg mouse model of AD alters core AD-related transcriptomic pathways in the hippocampus, and that it concomitantly restores essential components of hippocampal function, including dendritic spines and structure, immediate-early gene expression, and learning and memory itself. Moreover, SCD inhibition dampens activation of microglia, key mediators of spine loss during AD and the main immune cells of the brain. These data reveal that brain fatty acid metabolism links AD genes to downstream immune, synaptic, and functional impairments, identifying SCD as a potential target for AD treatment.

Alzheimer’s disease (AD) is characterized by lipid abnormalities which are not well understood. Here, the authors investigate the role of Stearoyl-CoA desaturase (SCD) in a mouse model of AD. They show that inhibiting SCD activity induces major brain and immune cell transcriptional changes and restores dendritic structure and learning and memory.

Applications and Techniques for Fast Machine Learning in Science

In this community review report, we discuss applications and techniques for fast machine learning (ML) in science-the concept of integrating powerful ML methods into the real-time experimental data processing loop to accelerate scientific discovery. The material for the report builds on two workshops held by the Fast ML for Science community and covers three main areas: applications for fast ML across a number of scientific domains; techniques for training and implementing performant and resource-efficient ML algorithms; and computing architectures, platforms, and technologies for deploying these algorithms. We also present overlapping challenges across the multiple scientific domains where common solutions can be found. This community report is intended to give plenty of examples and inspiration for scientific discovery through integrated and accelerated ML solutions. This is followed by a high-level overview and organization of technical advances, including an abundance of pointers to source material, which can enable these breakthroughs.

Immune phenotypic linkage between colorectal cancer and liver metastasis

The tumor microenvironment (TME) is connected to immunotherapy responses, but it remains unclear how cancer cells and host tissues differentially influence the immune composition within TME. Here, we performed single-cell analyses for autologous samples from liver metastasized colorectal cancer to disentangle factors shaping TME. By aligning CD45⁺ cells across different tissues, we classified exhausted CD8⁺ T cells (Texs) and activated regulatory T cells as M-type, whose phenotypes were associated with the malignancy, while natural killer and mucosal-associated invariant T cells were defined as N-type, whose phenotypes were associated with the niche. T cell receptor sharing between Texs in primary and metastatic tumors implicated the presence of common peripheral non-exhausted precursors. For myeloid cells, a subset of dendritic cells (DC3s) and SPP1⁺ macrophages were M-type, and the latter were predominant in liver metastasis, indicating its pro-metastasis role. Our analyses bridge immune phenotypes of primary and metastatic tumors, thereby helping to understand the tumor-specific contexture and identify the pro-metastasis components.

Longitudinal single-cell RNA-seq analysis reveals stress-promoted chemoresistance in metastatic ovarian cancer

Chemotherapy resistance is a critical contributor to cancer mortality and thus an urgent unmet challenge in oncology. To characterize chemotherapy resistance processes in high-grade serous ovarian cancer, we prospectively collected tissue samples before and after chemotherapy and analyzed their transcriptomic profiles at a single-cell resolution. After removing patient-specific signals by a novel analysis approach, PRIMUS, we found a consistent increase in stress-associated cell state during chemotherapy, which was validated by RNA in situ hybridization and bulk RNA sequencing. The stress-associated state exists before chemotherapy, is subclonally enriched during the treatment, and associates with poor progression-free survival. Co-occurrence with an inflammatory cancer-associated fibroblast subtype in tumors implies that chemotherapy is associated with stress response in both cancer cells and stroma, driving a paracrine feed-forward loop. In summary, we have found a resistant state that integrates stromal signaling and subclonal evolution and offers targets to overcome chemotherapy resistance.

scAnnotatR: framework to accurately classify cell types in single-cell RNA-sequencing data

BACKGROUND

Automatic cell type identification is essential to alleviate a key bottleneck in scRNA-seq data analysis. While most existing classification tools show good sensitivity and specificity, they often fail to adequately not-classify cells that are missing in the used reference. Additionally, many tools do not scale to the continuously increasing size of current scRNA-seq datasets. Therefore, additional tools are needed to solve these challenges.

RESULTS

scAnnotatR is a novel R package that provides a complete framework to classify cells in scRNA-seq datasets using pre-trained classifiers. It supports both Seurat and Bioconductor's SingleCellExperiment and is thereby compatible with the vast majority of R-based analysis workflows. scAnnotatR uses hierarchically organised SVMs to distinguish a specific cell type versus all others. It shows comparable or even superior accuracy, sensitivity and specificity compared to existing tools while being able to not-classify unknown cell types. Moreover, scAnnotatR is the only of the best performing tools able to process datasets containing more than 600,000 cells.

CONCLUSIONS

scAnnotatR is freely available on GitHub ( https://github.com/grisslab/scAnnotatR ) and through Bioconductor (from version 3.14). It is consistently among the best performing tools in terms of classification accuracy while scaling to the largest datasets.

Integrating pathway knowledge with deep neural networks to reduce the dimensionality in single-cell RNA-seq data

Background

Single-cell RNA sequencing (scRNA-seq) data provide valuable insights into cellular heterogeneity which is significantly improving the current knowledge on biology and human disease. One of the main applications of scRNA-seq data analysis is the identification of new cell types and cell states. Deep neural networks (DNNs) are among the best methods to address this problem. However, this performance comes with the trade-off for a lack of interpretability in the results. In this work we propose an intelligible pathway-driven neural network to correctly solve cell-type related problems at single-cell resolution while providing a biologically meaningful representation of the data.

Results

In this study, we explored the deep neural networks constrained by several types of prior biological information, e.g. signaling pathway information, as a way to reduce the dimensionality of the scRNA-seq data. We have tested the proposed biologically-based architectures on thousands of cells of human and mouse origin across a collection of public datasets in order to check the performance of the model. Specifically, we tested the architecture across different validation scenarios that try to mimic how unknown cell types are clustered by the DNN and how it correctly annotates cell types by querying a database in a retrieval problem. Moreover, our approach demonstrated to be comparable to other less interpretable DNN approaches constrained by using protein-protein interactions gene regulation data. Finally, we show how the latent structure learned by the network could be used to visualize and to interpret the composition of human single cell datasets.

Conclusions

Here we demonstrate how the integration of pathways, which convey fundamental information on functional relationships between genes, with DNNs, that provide an excellent classification framework, results in an excellent alternative to learn a biologically meaningful representation of scRNA-seq data. In addition, the introduction of prior biological knowledge in the DNN reduces the size of the network architecture. Comparative results demonstrate a superior performance of this approach with respect to other similar approaches. As an additional advantage, the use of pathways within the DNN structure enables easy interpretability of the results by connecting features to cell functionalities by means of the pathway nodes, as demonstrated with an example with human melanoma tumor cells.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13040-021-00285-4.

Temporal single-cell tracing reveals clonal revival and expansion of precursor exhausted T cells during anti-PD-1 therapy in lung cancer

Anti-PD-1 treatment has shown unprecedented clinical success in the treatment of non-small-cell lung cancer (NSCLC), but the underlying mechanisms remain incompletely understood. Here, we performed temporal single-cell RNA and paired T-cell receptor sequencing on 47 tumor biopsies from 36 patients with NSCLC following PD-1-based therapies. We observed increased levels of precursor exhausted T (Texp) cells in responsive tumors after treatment, characterized by low expression of coinhibitory molecules and high expression of GZMK. By contrast, nonresponsive tumors failed to accumulate Texp cells. Our data suggested that Texp cells were unlikely to be derived from the reinvigoration of terminally exhausted cells; instead, they were accumulated by (1) local expansion and (2) replenishment by peripheral T cells with both new and pre-existing clonotypes, a phenomenon we named clonal revival. Our study provides insights into mechanisms underlying PD-1-based therapies, implicating clonal revival and expansion of Texp cells as steps to improve NSCLC treatment.

Besca, a single-cell transcriptomics analysis toolkit to accelerate translational research

Single-cell RNA sequencing (scRNA-seq) revolutionized our understanding of disease biology. The promise it presents to also transform translational research requires highly standardized and robust software workflows. Here, we present the toolkit Besca, which streamlines scRNA-seq analyses and their use to deconvolute bulk RNA-seq data according to current best practices. Beyond a standard workflow covering quality control, filtering, and clustering, two complementary Besca modules, utilizing hierarchical cell signatures and supervised machine learning, automate cell annotation and provide harmonized nomenclatures. Subsequently, the gene expression profiles can be employed to estimate cell type proportions in bulk transcriptomics data. Using multiple, diverse scRNA-seq datasets, some stemming from highly heterogeneous tumor tissue, we show how Besca aids acceleration, interoperability, reusability and interpretability of scRNA-seq data analyses, meeting crucial demands in translational research and beyond.

Automatic cell type identification methods for single-cell RNA sequencing

Single-cell RNA sequencing (scRNA-seq) has become a powerful tool for scientists of many research disciplines due to its ability to elucidate the heterogeneous and complex cell-type compositions of different tissues and cell populations. Traditional cell-type identification methods for scRNA-seq data analysis are time-consuming and knowledge-dependent for manual annotation. By contrast, automatic cell-type identification methods may have the advantages of being fast, accurate, and more user friendly. Here, we discuss and evaluate thirty-two published automatic methods for scRNA-seq data analysis in terms of their prediction accuracy, F1-score, unlabeling rate and running time. We highlight the advantages and disadvantages of these methods and provide recommendations of method choice depending on the available information. The challenges and future applications of these automatic methods are further discussed. In addition, we provide a free scRNA-seq data analysis package encompassing the discussed automatic methods to help the easy usage of them in real-world applications.

Multiomics Analysis Reveals the Prognostic Non-tumor Cell Landscape in Glioblastoma Niches

A comprehensive characterization of non-tumor cells in the niches of primary glioblastoma is not fully established yet. This study aims to present an overview of non-malignant cells in the complex microenvironment of glioblastoma with detailed characterizations of their prognostic effects. We curate 540 gene signatures covering a total of 64 non-tumor cell types. Cell type-specific expression patterns are interrogated by normalized enrichment score across four large gene expression profiling cohorts of glioblastoma with a total number of 967 cases. The glioblastoma multiforms (GBMs) in each cohort are hierarchically clustered into negative or positive immune response classes with significantly different overall survival. Our results show that astrocytes, macrophages, monocytes, NKTs, and MSC are risk factors, while CD8 T cells, CD8 naive T cells, and plasma cells are protective factors. Moreover, we find that the immune system and organogenesis are uniformly enriched in negative immune response clusters, in contrast to the enrichment of nervous system in positive immune response clusters. Mesenchymal differentiation is also observed in the negative immune response clusters. High enrichment status of macrophages in negative immune response clusters is independently validated by analyzing scRNA-seq data from eight high-grade gliomas, revealing that negative immune response samples comprised 46.63 to 55.12% of macrophages, whereas positive immune response samples comprised only 1.70 to 8.12%, with IHC staining of samples from six short-term and six long-term survivors of GBMs confirming the results.

Integrating multiple references for single-cell assignment

Efficient single-cell assignment is essential for single-cell sequencing data analysis. With the explosive growth of single-cell sequencing data, multiple single-cell sequencing data sources are available for the same kind of tissue, which can be integrated to further improve single-cell assignment; however, an efficient integration strategy is still lacking due to the great challenges of data heterogeneity existing in multiple references. To this end, we present mtSC, a flexible single-cell assignment framework that integrates multiple references based on multitask deep metric learning designed specifically for cell type identification within tissues with multiple single-cell sequencing data as references. We evaluated mtSC on a comprehensive set of publicly available benchmark datasets and demonstrated its state-of-the-art effectiveness for integrative single-cell assignment with multiple references.

SSBER: removing batch effect for single-cell RNA sequencing data

BACKGROUND

With the continuous maturity of sequencing technology, different laboratories or different sequencing platforms have generated a large amount of single-cell transcriptome sequencing data for the same or different tissues. Due to batch effects and high dimensions of scRNA data, downstream analysis often faces challenges. Although a number of algorithms and tools have been proposed for removing batch effects, the current mainstream algorithms have faced the problem of data overcorrection when the cell type composition varies greatly between batches.

RESULTS

In this paper, we propose a novel method named SSBER by utilizing biological prior knowledge to guide the correction, aiming to solve the problem of poor batch-effect correction when the cell type composition differs greatly between batches.

CONCLUSIONS

SSBER effectively solves the above problems and outperforms other algorithms when the cell type structure among batches or distribution of cell population varies considerably, or some similar cell types exist across batches.

Knowledge-based classification of fine-grained immune cell types in single-cell RNA-Seq data

Single-cell RNA sequencing (scRNA-Seq) is an emerging strategy for characterizing immune cell populations. Compared to flow or mass cytometry, scRNA-Seq could potentially identify cell types and activation states that lack precise cell surface markers. However, scRNA-Seq is currently limited due to the need to manually classify each immune cell from its transcriptional profile. While recently developed algorithms accurately annotate coarse cell types (e.g. T cells versus macrophages), making fine distinctions (e.g. CD8+ effector memory T cells) remains a difficult challenge. To address this, we developed a machine learning classifier called ImmClassifier that leverages a hierarchical ontology of cell type. We demonstrate that its predictions are highly concordant with flow-based markers from CITE-seq and outperforms other tools (+15% recall, +14% precision) in distinguishing fine-grained cell types with comparable performance on coarse ones. Thus, ImmClassifier can be used to explore more deeply the heterogeneity of the immune system in scRNA-Seq experiments.

Accurate feature selection improves single-cell RNA-seq cell clustering

Cell clustering is one of the most important and commonly performed tasks in single-cell RNA sequencing (scRNA-seq) data analysis. An important step in cell clustering is to select a subset of genes (referred to as 'features'), whose expression patterns will then be used for downstream clustering. A good set of features should include the ones that distinguish different cell types, and the quality of such set could have a significant impact on the clustering accuracy. All existing scRNA-seq clustering tools include a feature selection step relying on some simple unsupervised feature selection methods, mostly based on the statistical moments of gene-wise expression distributions. In this work, we carefully evaluate the impact of feature selection on cell clustering accuracy. In addition, we develop a feature selection algorithm named FEAture SelecTion (FEAST), which provides more representative features. We apply the method on 12 public scRNA-seq datasets and demonstrate that using features selected by FEAST with existing clustering tools significantly improve the clustering accuracy.

A pan-cancer single-cell transcriptional atlas of tumor infiltrating myeloid cells

Tumor-infiltrating myeloid cells (TIMs) are key regulators in tumor progression, but the similarity and distinction of their fundamental properties across different tumors remain elusive. Here, by performing a pan-cancer analysis of single myeloid cells from 210 patients across 15 human cancer types, we identified distinct features of TIMs across cancer types. Mast cells in nasopharyngeal cancer were found to be associated with better prognosis and exhibited an anti-tumor phenotype with a high ratio of TNF+/VEGFA+ cells. Systematic comparison between cDC1- and cDC2-derived LAMP3+ cDCs revealed their differences in transcription factors and external stimulus. Additionally, pro-angiogenic tumor-associated macrophages (TAMs) were characterized with diverse markers across different cancer types, and the composition of TIMs appeared to be associated with certain features of somatic mutations and gene expressions. Our results provide a systematic view of the highly heterogeneous TIMs and suggest future avenues for rational, targeted immunotherapies.

Dissecting human embryonic skeletal stem cell ontogeny by single-cell transcriptomic and functional analyses

Human skeletal stem cells (SSCs) have been discovered in fetal and adult long bones. However, the spatiotemporal ontogeny of human embryonic SSCs during early skeletogenesis remains elusive. Here we map the transcriptional landscape of human limb buds and embryonic long bones at single-cell resolution to address this fundamental question. We found remarkable heterogeneity within human limb bud mesenchyme and epithelium, and aligned them along the proximal–distal and anterior–posterior axes using known marker genes. Osteo-chondrogenic progenitors first appeared in the core limb bud mesenchyme, which give rise to multiple populations of stem/progenitor cells in embryonic long bones undergoing endochondral ossification. Importantly, a perichondrial embryonic skeletal stem/progenitor cell (eSSPC) subset was identified, which could self-renew and generate the osteochondral lineage cells, but not adipocytes or hematopoietic stroma. eSSPCs are marked by the adhesion molecule CADM1 and highly enriched with FOXP1/2 transcriptional network. Interestingly, neural crest-derived cells with similar phenotypic markers and transcriptional networks were also found in the sagittal suture of human embryonic calvaria. Taken together, this study revealed the cellular heterogeneity and lineage hierarchy during human embryonic skeletogenesis, and identified distinct skeletal stem/progenitor cells that orchestrate endochondral and intramembranous ossification.

Benchmarking algorithms for pathway activity transformation of single-cell RNA-seq data

Biological pathway analysis provides new insights for cell clustering and functional annotation from single-cell RNA sequencing (scRNA-seq) data. Many pathway analysis algorithms have been developed to transform gene-level scRNA-seq data into functional gene sets representing pathways or biological processes. Here, we collected seven widely-used pathway activity transformation algorithms and 32 available datasets based on 16 scRNA-seq techniques. We proposed a comprehensive framework to evaluate their accuracy, stability and scalability. The assessment of scRNA-seq preprocessing showed that cell filtering had the less impact on scRNA-seq pathway analysis, while data normalization of sctransform and scran had a consistent well impact across all tools. We found that Pagoda2 yielded the best overall performance with the highest accuracy, scalability, and stability. Meanwhile, the tool PLAGE exhibited the highest stability, as well as moderate accuracy and scalability.

Adaptive immune responses to SARS-CoV-2 infection in severe versus mild individuals

The global Coronavirus disease 2019 (COVID-19) pandemic caused by SARS-CoV-2 has affected more than eight million people. There is an urgent need to investigate how the adaptive immunity is established in COVID-19 patients. In this study, we profiled adaptive immune cells of PBMCs from recovered COVID-19 patients with varying disease severity using single-cell RNA and TCR/BCR V(D)J sequencing. The sequencing data revealed SARS-CoV-2-specific shuffling of adaptive immune repertories and COVID-19-induced remodeling of peripheral lymphocytes. Characterization of variations in the peripheral T and B cells from the COVID-19 patients revealed a positive correlation of humoral immune response and T-cell immune memory with disease severity. Sequencing and functional data revealed SARS-CoV-2-specific T-cell immune memory in the convalescent COVID-19 patients. Furthermore, we also identified novel antigens that are responsive in the convalescent patients. Altogether, our study reveals adaptive immune repertories underlying pathogenesis and recovery in severe versus mild COVID-19 patients, providing valuable information for potential vaccine and therapeutic development against SARS-CoV-2 infection.

An entropy-based metric for assessing the purity of single cell populations

Single-cell RNA sequencing (scRNA-seq) is a versatile tool for discovering and annotating cell types and states, but the determination and annotation of cell subtypes is often subjective and arbitrary. Often, it is not even clear whether a given cluster is uniform. Here we present an entropy-based statistic, ROGUE, to accurately quantify the purity of identified cell clusters. We demonstrate that our ROGUE metric is broadly applicable, and enables accurate, sensitive and robust assessment of cluster purity on a wide range of simulated and real datasets. Applying this metric to fibroblast, B cell and brain data, we identify additional subtypes and demonstrate the application of ROGUE-guided analyses to detect precise signals in specific subpopulations. ROGUE can be applied to all tested scRNA-seq datasets, and has important implications for evaluating the quality of putative clusters, discovering pure cell subtypes and constructing comprehensive, detailed and standardized single cell atlas.

clustifyr: an R package for automated single-cell RNA sequencing cluster classification

Assignment of cell types from single-cell RNA sequencing (scRNA-seq) data remains a time-consuming and error-prone process. Current packages for identity assignment use limited types of reference data and often have rigid data structure requirements. We developed the clustifyr R package to leverage several external data types, including gene expression profiles to assign likely cell types using data from scRNA-seq, bulk RNA-seq, microarray expression data, or signature gene lists. We benchmark various parameters of a correlation-based approach and implement gene list enrichment methods. clustifyr is a lightweight and effective cell-type assignment tool developed for compatibility with various scRNA-seq analysis workflows. clustifyr is publicly available at https://github.com/rnabioco/clustifyr

clustifyr: an R package for automated single-cell RNA sequencing cluster classification

Assignment of cell types from single-cell RNA sequencing (scRNA-seq) data remains a time-consuming and error-prone process. Current packages for identity assignment use limited types of reference data and often have rigid data structure requirements. We developed the clustifyr R package to leverage several external data types, including gene expression profiles to assign likely cell types using data from scRNA-seq, bulk RNA-seq, microarray expression data, or signature gene lists. We benchmark various parameters of a correlation-based approach and implement gene list enrichment methods. clustifyr is a lightweight and effective cell-type assignment tool developed for compatibility with various scRNA-seq analysis workflows. clustifyr is publicly available at https://github.com/rnabioco/clustifyr.